Machine Dream - Continuous machine cognition research platform with GRASP loop, dreaming consolidation, and persistent memory. Pure LLM Sudoku solver that learns from experience and consolidates knowledge through AI "sleep cycles."

Brain-inspired cognitive architecture implementing basal ganglia RL, hippocampal memory consolidation, and prefrontal meta-cognition. Multi-agent system with dynamic attention control, procedural learning, and theory of mind - genuine cognitive continuity beyond context windows.

Deep learning implementation of the Benna-Fusi complex synapse model.

OpenDream — open-source memory consolidation for AI agents. Reads past any LLMs sessions, dreams across them to extract patterns, and writes consolidated memory into AGENTS.md so the next session starts smarter. Local SQLite, BYO LLM, no SaaS.

Self-hosted long-term memory substrate for any LLM-using tool. Hybrid vector + lexical recall with MMR reranking, auto-extracted knowledge graph, sleep-style consolidation, per-project tuning, MCP server, REST + OAuth 2.0, embedded React admin UI. SQLite or Postgres + pgvector or Qdrant. Pluggable providers (OpenAI, Anthropic, Gemini, Ollama).

adaptive memory layer for agentic AI

Dreaming plugin for OpenClaw memory-lancedb — Light/REM/Deep sleep phases with narrative generation and memory promotion.

Dream Memory, The persistent, reflective memory layer for AI systems. 4-phase consolidation loop (Orient→Gather→Consolidate→Prune) inspired by state-of-the-art LLM architectures.

Two paired Claude Code skills for safe memory consolidation — read-only walk + reflective synthesis + HTML approval gate. Inspired by leaked autoDream.

Testing whether selective forgetting improves LLM agent coherence over long conversations.

Offline memory consolidation for AI agents. Inspired by how brains dream.

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

An experimental topological field-memory prototype with vortex encoding and self-repairing field dynamics. It delivers noise-robust AI long-term memory as a complementary alternative to conventional RAG retrieval.

AI-powered memory consolidation for Claude Code — n8n workflow

Cognitive architecture for LLM chatbots with 4 memory systems and a consolidation process - semantic RAG, episodic recall, procedural rules, working memory, and periodic consolidation. Produces personalized responses from past conversations, learned behaviors, and document knowledge.

🧠 AI Agent Memory Layer — 双层记忆架构，让LLM Agent拥有持久记忆。支持语义搜索、记忆巩固、防污染检测。每天在生产环境运行。

Simulates distinct learning dynamics and synaptic stability, highlighting the role of axon-carrying dendrites (AcD) in memory consolidation under inhibitory conditions.